One of the big projects at the Steward Observatory Mirror Lab that is taking up most of my effort currently is the Large Synoptic Survey Telescope. This will be quite an amazing telescope - a huge (almost 28-foot-diameter or 8.4 meters) telescope with a wide field of view (7 full Moon diameters) that will image the entire sky in just a few days looking for asteroids, comets, variable stars and galaxies - most anything that goes bump in the night!
While I've blogged about the project before (hit the Steward Observatory Mirror Lab keyword down on the right side), not since we started our current task - diamond generating of the front surface of the mirror. It is sort of cool to perform a precision machining operation of such a large surface using diamond tools. And of course, it starts with the generating machine itself. The Large Optical Generator (LOG) was originally installed a couple blocks from here at the Optical Sciences Center's optics shop, where I started my career in optical fabrication in 1983. Two decades ago it was refurbished and moved to it's present location under the University of Arizona's football stadium east stands where the Mirror Lab resides. Mostly the LOG consists of a rotary table with 2-meter crossed roller bearings, and the main bridge, with horizontal and vertical carriage motions with a couple micron (thousandths of a millimeter) accuracy. The LSST mirror and it's cell weight something in excess of 40 tons, no problem for this machine.
This is a view looking down the ventilation system. The glass of the concave surface of LSST is at bottom. A large amount of coolant is used to keep the diamonds cool during the machining operation, some of which is turned into a mist by the rapid motion of the diamonds. Glass particles are carried away in the coolant and mist and because of the danger in ingesting the mist into our lungs, it is carried away by a ventilation system to our roof where it is filtered. The plastic flaps maintain a good seal to the glass and the silver tube in back is where the mist travels to the roof. The diamond wheel at center is turned by a 15 horsepower motor and there are 2 coolant lines visible, as well as an additional one with coolant running through the wheel itself. The gold-colored pellets are made of a soft metal with embedded diamonds which does the glass removal. The white football-shaped roller at right is a "safety roller" which rolls on the glass and monitors the glass position during the process.
The diamond pellets currently in use have 40 mesh diamonds - in other words, the diamonds are about 1/40th of an inch across, which are used for coarse generating. It results in a pretty rough surface, but we've got a few tons of glass to remove! With our standard generating process, we remove about 7 cubic inches of glass per minute, or about 2 cubic centimeters per second. You can see why we dump 15 gallons of coolant/minute on the diamond-glass interface. The coolant, by the way, is recycled and run through a centrifugal filter to remove the ground glass. In the closeup picture of the diamond pellet the diamonds are seen peeking from the metal matrix. As the pellets machine away the glass, the metal also wears exposing new diamonds, and releasing the dull ones as they wear. The few concave spots are where former diamonds were held, but were released as they dulled.
From the operator's platform Leslie keeps an eye on the operation. The ventilation system is visible at left. The actual generator action is enclosed by the mist containment, but a close watch is kept on the computer readouts and the mirror surface itself for problems. The yellow structure is a safety rail while we are working on the mirror surface. The steeply curved surface of LSST can be seen - the surface we are making has a 19 meter radius, or 8.5 meter focal length.
At right, grooves can be seen in the surface - the machining geometry we use produces a shallow groove, and as the mirror rotates, the diamond wheel spirals in and down towards the center, analogous to the grooves on a record. the black stripe was put on before the generating pass was started, and is removed as .030" is machined off this pass.
We've got quite a bit more to remove, so will be doing this for a while!
Monday, February 1, 2010
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment